With NIDA Behavioral Treatment Development Stage I and II support, researchers at the Center for Addiction and Pregnancy have developed and tested an intensive outpatient relapse-prevention approach, Reinforcement Based Treatment (RBT), for drug-addicted pregnant women. RBT is based on the principles of social learning theory and employs abstinence-contingent access to life-skills training, recreational therapy, and employment preparation and placement as routine aspects of an intensive day treatment counseling program. RBT has been tested in four randomized controlled trials4-7, with the most recent trial in pregnant patients who did not qualify for or refused methadone treatment as an adjunct to care. Results of this latter trial indicated that RBT participants, as compared to CAP standard care participants, had significantly greater CAP treatment utilization (61% v. 21%) at 3 months and significantly greater rates of abstinence from heroin (52% v. 29%; 67% v. 20%) and cocaine (48% v. 13%; 56% v. 20%) at 3 and 6 months, respectively, post-study entry. Objectives: This revised application is a Stage II behavioral development study designed to answer remaining critical questions necessary before disseminating RBT to the larger treatment community. These questions focus on the levels of intensity of RBT most efficacious for drug-addicted pregnant patients, the individual mediators and moderators that determine which RBT treatment level is best for which patient, and the cost-effectiveness of RBT treatment. Design: The proposed study utilizes a novel approach to conducting a controlled clinical trial, the sequential multiple assignment randomized trial (SMART) design. Participants (N=300) will first be randomized at treatment outset into either treatment-as-usual RBT or a reduced intensity RBT. All participants will receive a subsequent randomization based upon an assessment of their initial two weeks of treatment compliance. Early-non-compliant participants will be randomized to receive either the same or an increased level of RBT treatment intensity while early-compliant participants will be randomized to receive either the same or decreased level of treatment intensity and scope. Primary outcome measures include treatment completion and maternal heroin and cocaine use. Secondary outcome measures include maternal measures of HIV risk behavior, other drug use, and psychosocial functioning and neonatal measures of length of hospitalization, physical birth parameters, neurological integrity, and behavioral functioning. Innovation and Significance: The proposed project's innovation includes: the novelty RBT, use of a cutting- edge SMART model, application of advanced statistical techniques and inclusion of a cost-effectiveness approach. The proposed project's significance is exceedingly high, as it will lay the foundation for later Stage III studies focused on dissemination of stepped care treatment programs for drug-addicted pregnant women that can be implemented not only in comprehensive care clinics but in diverse community settings that provide services to such women.